network

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It exhaustively details the deterministic risk dictionary, the zero-LLM/zero-guessing policy, the handling of missing targets (null fields), risk firing only on parse/regex evidence, the redact behavior, limit bounds with truncated flag, tri-state is_error semantics, and the error contract mapping. Honesty caveats are explicitly documented, such as risk labels not being a threat oracle.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but every sentence earns its place for an 11-parameter audit tool. It is front-loaded with a clear one-sentence summary, then structured into how it works, filters, output, and honesty caveats. No filler or redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (11 parameters, 0% schema description coverage, no annotations) and the presence of an output schema, the description is remarkably complete. It covers the full output shape (count, risky_count, by_domain, by_risk, truncated, diagnostics), edge cases (empty results, missing targets), and error handling. An agent has everything needed to select and invoke the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, but the description adds meaningful semantics for every parameter: agent, session (uuid or list), since/until (ISO bounds on call ts), kind (fetch|search with fail-loud on unknown values), risk (include only/exclude), domain (equals-or-subdomain match), noise and project_dir (session-level semantics), limit (default 50, 0 = no cap), and redact. This goes far beyond the raw schema types and defaults.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool performs a Network-egress audit, answering 'where did an agent reach out to the network — and how risky did those requests look?'. It distinguishes itself from siblings by explaining it is a preset over the existing core (not a second engine) and focuses specifically on web tool calls (WebFetch, WebSearch, webfetch, web_search, etc.).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-to-use context: one call answers network egress and risk. It also gives clear when-not-to-use guidance, noting MCP-mediated network access stays under tool_kind='mcp' and is never guessed into this audit, and that Pi records no web tool (honest absence). It references query/read_session for walking context, effectively suggesting alternatives for further investigation.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.